Six checks on one specific deal: the terms, the cap table ahead of you, what your stake becomes after several more rounds of nothing going wrong, the preference stack, the traction evidence, and whether you have reserved capital to defend the position at all.
14 min read · Level 3
The check I would put first if I could only keep one is the ordinary future-dilution calculation, and I think it is the one most often skipped precisely because it requires no bad news to produce an uncomfortable answer. Nobody has to be wrong, nobody has to make a mistake, and no red flag has to appear in a data room for a 5.0% stake to become 2.56% - it only requires the company to do the thing that a successful early-stage company is supposed to do, which is raise more money at higher valuations from new investors who insist on their own slice of the pie. An investor who only ever prices a deal against today's cap table has quietly agreed to be surprised later by an outcome that was computable, in full, on the day the term sheet was signed. What I would ask any new angel to internalise is that evaluating a specific opportunity in this asset class is not primarily a judgement about the founders or the market - those matter, and check one and check six speak to them - but is, underneath that, an exercise in reading a legal document accurately enough to know what you actually bought, what it is likely to become through no fault of anyone's, and what it would take, in capital and attention, to defend it once it starts working. Most of the work in this lesson is arithmetic that anyone could do. Almost nobody does it before signing.
Six checks on one specific opportunity, ordered by how much each one changes the eventual outcome rather than by how quickly it can be answered.
One. What the company actually does, in one plain sentence, stated to someone outside the industry. If a genuine attempt cannot produce that sentence, the gap is the finding, not a sign that you need more technical background.
Two. The instrument and its terms. Which of lesson one's three arrangements is being offered, the cap and discount or the round price, the liquidation preference multiple, whether it participates, and whether an anti-dilution provision protects the investor's price if a future round prices lower - a down round.
Three. The cap table ahead of you. How large the option pool is, how many prior SAFEs or notes are stacked and on what terms, and what percentage the round you are being offered actually represents once every prior instrument has converted - not before.
Four. What your stake becomes after ordinary future dilution, not just today. This is the calculation most first-time angels never run, and it needs no bad outcome to matter. A typical seed check might represent 5.0% ownership at close. Three further priced rounds, each diluting existing holders by a representative 20% through new investor shares and option pool top-ups, leave the position at:
After Series A 5.0% x 0.8 = 4.0% After Series B 4.0% x 0.8 = 3.2% After Series C 3.2% x 0.8 = 2.56%
A nominal 5.0% stake becomes 2.56% - a cumulative reduction of 48.8% - purely from three rounds of ordinary financing in which nothing went wrong and the company may have executed flawlessly throughout. On a 200,000,000 eventual exit, the difference between the nominal and the diluted stake is 10,000,000 against 5,120,000: a 4,880,000 gap that has nothing to do with the company's performance and everything to do with an arithmetic process that happens to every investor who does not defend against it.
Five. The preference stack ahead of any common or earlier, more junior stock. Each round typically layers its own preference on top of the ones before it. A seed round at a 2,000,000 preference, a Series A at 2,000,000, and a Series B at 5,000,000 together create a 9,000,000 stack that must be satisfied - in the ordinary seniority order, most senior last-money-in first - before common or subordinate holders see anything. A company that sells for exactly 9,000,000 returns the entire proceeds to preferred holders and nothing at all to common, without a single term being unusual.
Six. The evidence behind the traction claims, and your own capacity to follow on. Customer concentration, the honesty of any revenue figures, and whether you have reserved the capital pro-rata rights would require you to deploy if this position starts working.
Because checks three and four are the ones almost nobody runs, and they are frequently the ones that decide the eventual outcome regardless of how the company itself performs.
Check three matters because a round's stated percentage is only ever accurate against the share count it was calculated on, and that share count is set by the company's own counsel from a cap table the investor rarely sees in full before committing. A round "selling 10.0% for 2,000,000" calculated against a share count that omits an already-agreed but unissued option pool top-up is selling something smaller than 10.0% the moment that pool is actually created - a detail buried in the term sheet's definitions section rather than its headline terms.
Check four is arithmetic, not judgement, and that is exactly why it is so often skipped: it produces an uncomfortable number about a company that may be doing everything right. The 48.8% cumulative reduction in the worked example required no misstep by the founders, no missed milestone, and no bad luck - it required only three more financings of the ordinary kind that a successful, growing company raises as a matter of course. An investor who evaluates a deal purely on today's ownership percentage and today's exit-scenario math has evaluated a number that will very likely be smaller, sometimes by nearly half, by the time an exit actually occurs.
The preference stack in check five compounds with check four rather than sitting apart from it: as more rounds are raised, both the investor's percentage shrinks and the amount that must be paid out ahead of common grows. A seed investor evaluating a deal purely on the company's current, modest preference stack is evaluating a stack that will almost certainly be larger, and senior to their own position, by the time of an exit several rounds later.
And check six's traction evidence deserves the same discipline the-information-request and verification-vs-assumption apply everywhere else in this curriculum: a customer count, a revenue figure, or a growth rate stated by the founders in a pitch deck is a claim, not a verified fact, and the private-market absence of any regulatory filing requirement means nobody but the investor is checking it before the money moves.
This is educational information, not investment, securities or financial advice. Rules differ by country and change — get advice from a qualified professional in your jurisdiction before acting.
Educational, not personalised financial advice. Nothing in the Investor Academy is a recommendation to buy, sell or hold anything — what is right for you depends on your own circumstances and the rules where you live.
Sources: U.S. SEC - Investor.gov, protecting your investments; FINRA - investor information
Two seed investments, both offered at an identical 5.0% for 250,000, implying a 5,000,000 post-money valuation, evaluated against the six checks.
Company D. The cap table shared before closing shows a 15% unallocated option pool that has not yet been created, meaning the effective post-money once it is carved out is closer to 5,882,353 (5,000,000 / 0.85), and the true ownership purchased is closer to 4.25% than the stated 5.0%. The preference is a standard 1x non-participating. Prior SAFEs total 400,000 on reasonable terms. Customer references confirm the stated 40 paying customers independently, with no single customer above 15% of revenue.
Company E. Identical headline terms - 5.0% for 250,000 - but the option pool question goes unanswered by the founders when asked directly, twice. The preference is 1.5x participating, a materially more investor-protective and more senior structure than standard. Three prior SAFEs, totalling 600,000, carry uncapped, discount-only terms - meaning their eventual dilutive effect on this round cannot be computed in advance. The stated 40 customers cannot be corroborated; two reference calls go unanswered and the founders decline a third.
Running the ordinary three-round dilution calculation from check four on both: Company D's realistic 4.25% becomes roughly 2.176% after three further 20% dilutive rounds (4.25% x 0.8^3), while Company E's stated 5.0% - already suspect - would become 2.56% under the same assumption, but the uncapped prior SAFEs mean the true starting percentage, and therefore the true ending one, cannot actually be computed with the information available.
Company D fails no check outright but reveals a real, quantifiable 15.0% gap between the headline and effective ownership - a fact worth negotiating on, not a reason to walk away. Company E fails checks two, three, five and six on evidence available before any money moves: an unanswerable pool question, an unusually senior preference, an uncomputable prior-dilution position, and uncorroborated traction claims. Neither failure required a forecast about the company's eventual success or failure - only the six checks, run in order, before committing.
Evaluating a deal on its headline ownership percentage without confirming the option pool has already been carved out of the pre-money share count. An unallocated pool created after the round dilutes the new investor along with everyone else already on the table, and it is routinely negotiated to be carved out before the round closes specifically so that it does not.
Skipping the ordinary future-dilution calculation because the company looks strong today. It is arithmetic that applies regardless of the company's quality, and it is the single most reliably skipped step in this entire evaluation - a 5.0% stake becoming 2.56% after three unremarkable further rounds needs no misstep by anyone to occur.
Accepting a materially more senior or participating preference structure without recognising what it costs junior holders, including, eventually, this same investor once further rounds are raised above them. A term that looks like it only protects you as the current investor becomes a term working against you the moment you are no longer the most recent money in.
Treating founder-supplied traction figures as verified. A customer count or revenue figure in a pitch deck is a claim; the-information-request and verification-vs-assumption both apply in full here, and a private company faces no regulatory filing requirement forcing independent verification the way a public one does.
Committing the initial cheque without reserving capital for a follow-on. A deal that looks correct on every other check but leaves the investor unable to exercise pro-rata rights if the company starts working has been evaluated against only half of what the position will eventually require.
No industry jargon, stated to someone outside the sector. If it cannot be written after a genuine attempt with the pitch materials open, treat that as the finding rather than a personal knowledge gap.
Ask directly and in writing. An unallocated pool created after your round dilutes you along with the founders, and the gap between the stated and effective percentage can run into several points on an otherwise ordinary deal.
Every SAFE, every note, its cap, its discount, and whether it is priced pre-money or post-money. An uncapped prior instrument makes your own eventual ownership uncomputable until it converts, which is itself worth knowing before you commit.
Multiply by 0.8 for each plausible future priced round the company will need to raise before an exit. On a 5.0% stake, three rounds bring it to roughly 2.56% with nothing having gone wrong - compute this before comparing the deal's economics against any exit scenario.
Every prior and current round's preference, summed. A stack that already consumes a large share of a plausible modest exit is a red flag about the downside regardless of how promising the upside story sounds.
A reference call to an actual customer, a check of a stated partnership, or any verification that does not originate from the founders themselves. Two unanswered or declined reference requests is itself an answer.
If a follow-on round arrives in eighteen months and you cannot fund your pro-rata share, decide now whether that is an acceptable outcome for this specific position, rather than discovering the constraint at the moment the company is finally proving itself.
Before committing to any specific seed or early-stage opportunity, when re-evaluating an existing position ahead of a follow-on decision, and whenever a deal is being recommended primarily on the strength of the team or the story rather than on terms that can be checked.
Do not use these six checks to rank several unrelated companies against each other on a single score - they produce a description of one deal's terms and risks, not a comparable index across companies in different sectors and stages. And do not treat passing every check as a reason the company will succeed; the checks establish what you actually own and what it could become, not whether the underlying business will work.
A newly created or expanded option pool ahead of a priced round is a completely ordinary term, and it is also one of the most common sources of a gap between a round's headline percentage and what an investor actually ends up owning. The pool is frequently discussed verbally as though it were already accounted for in the pre-money figure everyone is negotiating around, and the actual mechanics of when and how it is carved out are settled later, in definitions buried well past the term sheet's summary page. By the time the final documents are signed, the investor's true ownership can be a meaningful step below the number that was quoted throughout every earlier conversation, without anyone having said anything untrue at any point.
Ask, in writing, whether the option pool is being carved out of the pre-money or the post-money before agreeing to any percentage. The two produce different effective ownership from an identical headline number.
Create a free account to read the rest and apply it to your own investing.